Refusals: BioTIER
Can the model reason accurately about difficult scientific material?
Science5 ranked models5 reported values1 reportsContextual score
Top rankings
One row per model, using its best reported score across effort settings.
| Rank | Model | Best score | Best reported setting |
|---|---|---|---|
| 1 | Muse Spark 1.0Meta | 98.0% | Reported configurationMeta report1 value · 1 reportJul 9, 2026 · source |
| 2 | Muse Spark 1.1Meta | 97.7% | xhigh effortMeta report1 value · 1 reportJul 9, 2026 · source |
| 3 | Claude Opus 4.8Anthropic | 94.9% | max effortMeta report1 value · 1 reportJul 9, 2026 · source |
| 4 | GPT-5.5OpenAI | 69.2% | xhigh effortMeta report1 value · 1 reportJul 9, 2026 · source |
| 5 | Gemini 3.1 ProGoogle | 57.7% | high effortMeta report1 value · 1 reportJul 9, 2026 · source |
Effort curve
Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarylimited methodology
Can the model reason accurately about difficult scientific material?
Scientific knowledge and reasoning under the benchmark's published question set. The source reports a refusal percentage; interpret with the named safety objective.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Method / source